1. International Journal of Obesity (2000) 24, 906±914
ß 2000 Macmillan Publishers Ltd All rights reserved 0307±0565/00 $15.00
www.nature.com/ijo
Appetite after weight loss by energy restriction
and a low-fat diet±exercise follow-up
 Á
E Doucet1, P Imbeault1, S St-Pierre1, N Almeras1, P Mauriege1, D Richard1 and A Tremblay1*
1
Â
Division of Kinesiology, Department of Social and Preventive Medicine, Laval University, Ste-Foy, Quebec, Canada G1K 7P4
OBJECTIVE: The aim of the present study was to determine the impact of weight loss on appetite as measured by
visual analog scale (VAS).
METHODS: Seventeen subjects (10 men and seven women) took part in a 15 week weight loss program which
consisted of drug therapy (fen¯uramine 60 mgaday) or placebo coupled to an energy restriction (À2930 kJaday; phase
1) followed by an 18 week low-fat diet ± exercise follow-up (phase 2). Subjects were given a standardized breakfast
before and after phase 1 as well as after phase 2. Individuals were asked to ®ll out VAS before and at 0, 10, 20, 30, 40,
50 and 60 min after this test meal. Blood samples were drawn before the meal and at 0, 30 and 60 min postprandially
and analyzed for glucose and insulin. Fasting plasma cortisol and leptin were also determined.
RESULTS: An increase in the fasting desire to eat, hunger and prospective food consumption (PFC) was observed after
phase 1 and to an even greater extent after phase 2 in both men and women. In the fasting state, positive correlations
were observed between changes in the desire to eat (r ˆ 0.76; P ` 0.05) as well as changes of PFC (r ˆ 0.82; P ` 0.05)
and changes in cortisol at the end of phase 1 for women. In response to phase 1, statistically signi®cant correlations
were found between changes of hunger (r ˆ 0.64; P ` 0.05) and desire to eat (r ˆ 0.67; P ` 0.05) as measured by AUC in
response to the meal and changes of fasting plasma cortisol in men. The most consistent predictor of changes of
baseline desire to eat (r ˆ 0.68 P ` 0.05), fullness (r ˆ À0.78, P ` 0.05) and PFC (r ˆ 0.91, P ` 0.01) during phase 2 was
the change in fasting cortisol in men. Changes of fullness were also associated with changes of fasting leptin in men
(r ˆ 0.68; P ` 0.05) during phase 2.
CONCLUSION: These results suggest that weight loss is accompanied by an increase of baseline appetite in both men
and women and that the most consistent predictor of these changes in appetite seems to be changes in fasting
plasma cortisol.
International Journal of Obesity (2000) 24, 906±914
Keywords: weight loss; appetite; cortisol; visual analog scale
Introduction etiology of obesity, and nutritional strategies are
now consequently being developed by health profes-
sionals to promote satiety with a minimal energy
Obesity prevalence has increased considerably intake.
throughout the latter half of this century. This increase Another important dimension of this problem that
in obesity was paralleled to important changes in food has been scarcely documented up to now is the
consumption patterns.1 In a context where virtually question as to whether appetite is altered by weight
unlimited access to high-fat foods is made possible, loss. This hypothesis came from the fact that weight
the maintenance of energy balance seems to be loss is characterized by numerous metabolic andaor
dif®cult to achieve for many individuals. This might endocrine adaptations that might in turn affect appe-
be due to the fact that this type of dietary regimen has tite in the reduced-obese state. Indeed, weight loss has
a weak potential to produce potent satiety signals and been shown to reduce sympathetic nervous system
to inhibit subsequent energy intake.2,3 This situation activity,4 plasma leptin5 and insulin,6 all of which have
often leads to passive overconsumption and thus to been shown to inhibit food and energy intake.7 ± 10
positive energy and fat balance, and over time to Moreover, weight loss has also been shown to
excess body fat storage. In this context, it can be improve glycemic control,11 which is partly re¯ected
speculated that the potential of this type of dietary by a decrease in fasting glucose. This might be
regimen to suppress appetite-related feelings may not problematic in a context where a decrease in glycemia
be as effective as that which characterized the lifestyle has been shown to trigger episodes of feeding.12,13 It is
of our ancestors. Thus, subjective feelings of hunger thus possible that, in order to achieve a comparable
and satiety might play an important role in the level of satiety to what was experienced while still
obese, a more substantial amount of energy has to be
consumed to compensate for the reduced impact of the
*Correspondence: A Tremblay, Division of Kinesiology, PEPS, above-mentioned factors on appetite-related variables.
Laval University, Ste-Foy, Quebec, Canada G1K 7P4.
 This phenomenon might in turn partly explain the fact
E-mail: angelo.tremblay@kin.msp.ulaval.ca
Received 23 July 1999; revised 17 December 1999; accepted that most reduced-obese individuals regain the weight
23 February 2000 lost within a relatively short period of time if no

2. Weight loss and appetite
E Doucet et al
907
intervention is undertaken to prevent this unfortunate valvular function,20,21 all subjects (including placebos)
outcome.14 were subjected to an echocardiogram. Following this
Few studies have directly addressed the issue as to assessment, a detailed analysis of cardiac valvular
whether appetite scores are affected by prolonged function was performed by cardiologists, who
energy de®cit and consequent weight loss. In this detected no abnormalities in response to the use of
context, Heini et al 15 have observed that during fen¯uramine under these conditions.22
controlled weight loss, although no signi®cant
changes in hunger±satiety levels were found, some
positive correlations were observed between postpran- Weight loss program
dial changes in these variables and fractional changes This program was designed as a two-phase program as
in glucose and insulin. Results from the same group previously described.19 Phase 1 consisted of a 15-
have also demonstrated that during energy restriction, week non-macronutrient speci®c energy restriction of
changes in leptin are accompanied by changes in 2930 kJaday (À700 kcaladay) coupled to drug therapy
hunger±satiety ratings16 which might be partly (fen¯uramine 60 mgaday) or placebo. Following
explained by the fact that leptin levels were shown phase 1, a low fat diet from which 30% of total
to predict palatability scores.17 It has also been energy intake (EI; 8445 Æ 158 and 7590 Æ 273 kJ for
reported that, in response to weight loss induced by men and women, respectively) came from fat, 53%
energy restriction and physical activity, changes in from carbohydrates and 17% from proteins, as well as
day-long visual analog scale (VAS) measurements an aerobic exercise prescription (60±75% of VO2
were associated with changes in leptin and glucose.18 max; 3±5 times a week; 45±60 minasession) were
We thus hypothesized that the reduced-obese state prescribed to subjects for a mean duration of 18 weeks
would be characterized by an increase of VAS vari- (phase 2). To ensure proper monitoring of exercise
ables measured in the fasting state and that a given intensity and duration and to verify the subjects'
amount of calories would exert a lesser in¯uence over compliance to the exercise prescription, they had to
the subjective feelings of hunger and satiety due to wear a heart rate monitor (Polar Vantage XLTM HRM,
some hormonal adaptations to weight loss. In this Stamford, CT) during their exercise sessions to assess
study, we measured the impact of a standardized mean heart rate and duration of the training sessions.
breakfast test meal on the subjective feelings of Noteworthy is the fact that phase 2 was continued
appetite as measured by VAS as well as their relation until a resistance to further lose fat was achieved in
to changes in fasting plasma glucose, insulin, leptin subjects. That is, a threshold beyond which a further
and cortisol in the obese state, after a 15-week drug reduction andaor manipulation of energy intake
therapy intervention in the reduced-obese state, and andaor an increase in energy expenditure derived
after an additional 18-week low-fat diet±exercise from physical activities would have caused subjects
follow-up. to feel unable to comply with the prescription over
time. This was done to ensure long term maintenance
of the healthier lifestyle which resulted from this
intervention program. Since this approach was indivi-
Methods dualized, some subjects reached this resistance to
further lose fat before others and thus the duration
of this follow-up period was different amongst indi-
Seventeen obese subjects (10 men and seven women) viduals (from 8 to 24 weeks).
underwent weight loss by drug therapy (60 mgaday of
fen¯uramine) or placebo treatment coupled to a non-
macronutrient speci®c energy restriction for a 15- Standardized breakfast test meal
week period (phase 1). Since this study was double- Subjects were asked to come to the laboratory after a
blind, subjects were given a number in the order in 12 h overnight fast. They were then asked to eat a
which they were recruited for the study. Drug or standardized breakfast meal test which consisted of
placebo had already been randomly assigned to whole wheat bread, butter, peanut butter, strawberry
these numbers while respecting the ratio of 5 to 1 jam, mozzarella cheese and orange juice. The meal
drug- or placebo-treated individual, respectively. was designed to have a food quotient of 0.85 and an
Phase 1 was followed by an $18-week low-fat energy content of 2993 kJ (715 kcal) and 2574 kJ
diet±exercise follow-up (phase 2). A detailed descrip- (615 kcal) for men and women, respectively (see
tion of this two-phase weight loss program is given Appendix A). To increase the energy content of the
below and has been previously described.19 Subjects test meal in men without changing the food quotient, a
gave their written consent to participate in this study, glass of 2% milk fat milk was added to this breakfast
which received approval from the Laval University test meal. Subjects were instructed to eat everything
Medical Ethics Committee. within a 20 min period. This test meal measurement
It is important to note that following the suspension was performed at three different times during the
of fen¯uramine and dexfen¯uramine further to a protocol, before phase 1 as well as after phases 1
potential association with disturbances in cardiac and 2.
International Journal of Obesity

3. Weight loss and appetite
E Doucet et al
908
Visual analog scale measurements (VAS) Glucose and insulin concentrations
Desire to eat, hunger, fullness and prospective food Blood samples were collected in tubes containing
consumption (PFC) were rated immediately before EDTA and Trasylol (Miles Pharmaceutics, Rexdale,
(after a 12 h overnight fast) and at 0, 10, 20, 30, 40, Ontario, Canada) in the fasting state at 0, 30 and
50 and 60 min after the standardized test meal on a 60 min after the meal. Plasma glucose was measured
150 mm VAS which was adapted from Hill and enzymatically,25 whereas plasma insulin concentra-
Blundell.23 Questions were asked as follows: (1) tion was determined by radioimmunoassay with poly-
`How strong is your desire to eat?' (very weak±very ethylene glycol separation.26
strong); (2) `How hungry do you feel?' (not hungry at
all±as hungry as I have ever felt); (3) `How full do
you feel?' (not full at all±very full); and (4) `How Plasma leptin and cortisol concentrations
much food do you think you could eat?' (nothing at Fasting plasma leptin concentrations were determined
all±a large amount). The before breakfast measure- with a highly sensitive commercial double-antibody
ment was considered as the fasting measurement for RIA (Human Leptin Speci®c RIA Kit, LINCO
all four variables. In order to calculate values of AUC Research, St Louis, MO, USA), which detects rela-
in response to the meal for the above variables, tively low leptin levels of 0.5 ngaml and which does
measurements at 0, 10, 20, 30, 40, 50 and 60 min not crossreact with human insulin, proinsulin, gluca-
were considered in this calculation by using the gon, pancreatic polypeptide or somatostatin. Our
trapezoid method, thus not including the fasting coef®cients of variation for the repeated assays
score obtained before breakfast. Hence, higher ranged from 4.0 to 5.5% for the lower leptin concen-
values for AUC in response to the meal are indicative trations and from 6.5 to 8.5% for higher plasma leptin
of a lesser suppression of the test meal for the desire to concentrations.27 Plasma cortisol concentrations were
eat, hunger and PFC and at the opposite greater values determined by radioimmunoassay (ICN Biomedicals
of AUC in response to the meal for fullness indicate a Inc., Costa Mesa, CA) from fasting plasma samples.
greater impact of the meal over this variable. Note-
worthy is the fact that the measurements of VAS were
performed at least 2±4 weeks after the interruption of Statistical analysis
drug therapy after phase 1 and at least 48 h after the Jump Sofware 3.1.6.2. from the SAS Institute Inc.
last bout of exercise following phase 2, when they had (Cary, NC, USA) was used for all analysis. Multi-
become resistant to further losses of body fat. At all variate analysis of variance (MANOVA) for repeated
sampling times subjects had been weight stable. In measures were ®rst performed on all variables to
this sense, we can consider that subjects were in assess the effects of gender, treatment, time and
energy balance even if weight stability is a gross their interaction. Since gender effects as well as
index of energy balance. Moreover, before every genderÂtime interactions were observed, genders
visit to the laboratory subjects were instructed to eat were analysed separately. Moreover, since no treat-
as they would normally. Women were also tested mentÂtime interactions were noted, placebos and
between days 5 and 12 of their menstrual cycle at drug treated individuals were pooled together for
all three testing periods. Also of importance is the fact further analysis. To verify the effect of treatment on
that VAS measurements were always performed in the VAS variables, paired t-tests were performed with a
same environment, ie at the same table with the same Bonnferroni correction. For these analyses, since two
lighting in the same room, which was kept free of comparisons were performed, the alpha level was set
odors and sounds as well as other factors that might at 0.025. Pearson correlations were also performed
contaminate this measurement (visual stimuli, indivi- between the changes of VAS variables (fasting and
duals in the room, etc). Under these conditions, VAS AUC in response to the test meal) as well as hormonal
measurements in our laboratory have been shown to changes. The alpha level for Pearson correlations was
be highly reliable both before and in response to a set at P ` 0.05. All data are expressed as mean
meal.24 Æ s.e.m.
Blood sampling and analysis
It is important to note that all blood samples were Results
drawn after an overnight fast of at least 12 h through a
venous catheter from an antecubital vein at around
08:00 at all three sampling times. Moreover, subjects As shown in Table 1, phase 1 caused a signi®cant
were instructed to abstain from physical exercise 48 h decrease in body weight (À11.6 and À7.3 kg), body
before blood sampling. Subjects were also instructed mass index (À3.8 andÀ2.9 kgam2) and fat mass (À9.9
to eat as they would normally before visits to the and À5.6 kg) in men and women respectively,
laboratory and women were tested between days 5 and whereas phase 2 caused a signi®cant decrease in fat
12 of their menstrual cycle before and after phase 1 as mass in men only (À4.1 kg). Fat-free mass was also
well as after phase 2. slightly reduced in response to phase 1 in both men
International Journal of Obesity

4. Weight loss and appetite
E Doucet et al
909
Table 1 Subjects' characteristics before and after treatment
Before phase1 After phase1 After phase 2
Variables Men Women Men Women Men Women
Age (y) 45.0 Æ 1.1 40.1 Æ 1.8 Ð Ð Ð Ð
Body weight (kg) 102.2 Æ 3.3 86.7 Æ 2.3 90.6 Æ 2.4a 79.4 Æ 3.2a 88.2 Æ 1.8a 77.6 Æ 3.3a
BMI (kgam2) 33.6 Æ 0.7 34.3 Æ 0.8 29.8 Æ 0.6a 31.4 Æ 1.2a 29.1 Æ 0.4a 30.1 Æ 1.3a
Fat mass (kg) 37.5 Æ 2.1 41.6 Æ 2.4 27.6 Æ 1.5a 36.0 Æ 2.4a 23.5 Æ 1.1b 33.4 Æ 3.1a
Fat-free mass (kg) 64.8 Æ 2.0 45.1 Æ 1.7 63.1 Æ 2.0 43.3 Æ 1.3 64.7 Æ 1.9 44.3 Æ 2.0
Mean Æ standard error of the mean; n ˆ 10 and n ˆ 7 for men and women, respectively; BMI ˆ body mass index;
means which do not share the same letter are signi®cantly different from one another.
and women and, as expected, a slight non-signi®cant hunger were signi®cantly greater than before phase 1
increase of this variable was noted after phase 2. values, while PFC was only signi®cantly different
from before phase 1 values in women. The measure-
Changes of fasting VAS scores in response to phases 1 ment of fasting fullness remained statistically
and 2 unchanged in response to this whole program, be it
Figure 1 presents results from the comparison of the after phase 1 or 2 in both sexes.M
effects of weight loss on VAS variables measured in
the fasting state in the 17 subjects who completed both Changes of post-prandial VAS scores in response to
phases 1 and 2. As a result of this program, the fasting phases 1 and 2
desire to eat, hunger and PFC were all increased in Values of AUC in response to the test meal for VAS
response to phase 1, although not signi®cantly. More- variables are shown in Table 2. No signi®cant differ-
over, at the end of phase 2 fasting desire to eat and ence was observed for both men and women be it after
Figure 1 Effects of weight loss on VAS variables measured in the fasting state before and after 15 weeks of drug therapy (phase 1) as
well as after an 18 week low-fat diet ± exercise follow-up (phase 2). *,**signi®cantly different from before phase 1 measurement at 0.03
and 0.01, respectively. PFC ˆ prospective food consumption and n ˆ 10 men and 7 women.
International Journal of Obesity

5. Weight loss and appetite
E Doucet et al
910
phases 1 or 2, indicating that a meal of a given caloric only variable that came close to statistical signi®cance
content seems to in¯uence variables of VAS in a was fullness (r ˆ 0.71, P ˆ 0.07).
similar fashion before and after weight loss. Simple correlations were also performed on
changes of VAS variables calculated as AUC in
response to the test meal which occurred during
Correlates of changes of VAS scores in the fasting and phase 2 of VAS and changes of fasting insulin,
postprandial states during phase 1 glucose, leptin and cortisol as well as AUC for
Simple correlations between changes in VAS vari- glucose and insulin measured during the test meal.
ables measured in the fasting state and changes in These analyses failed to reveal any signi®cant asso-
fasting insulin, glucose, leptin and cortisol which ciation (results not shown).
occurred during phase 1 were performed for both Figure 2 presents the regression lines for changes in
men and women. These analyses did not reveal any desire to eat and PFC in the fasting state plotted
signi®cant association, except for changes in cortisol against changes in fasting cortisol, insulin and
which were signi®cantly associated to changes in leptin. As expected, even when pooling men and
desire to eat (r ˆ 0.76, P ` 0.05) and to changes women together, the association was still statistically
in PFC (r ˆ 0.82, P ` 0.05) in the fasting state for signi®cant between fasting values of desire to eat
women. As for changes in the variables of VAS (r ˆ 0.58, P ` 0.01) and PFC (r ˆ 0.67, P ` 0.01)
calculated as AUC in response to the test meal, and fasting cortisol. On the other hand, no signi®cant
signi®cant associations were found between changes associations were found between these two markers of
in fasting cortisol and changes in AUC desire to eat appetite and fasting insulin and leptin.
(r ˆ 0.66, P ` 0.05) and hunger (r ˆ 0.64, P ` 0.05)
in men. On the other hand, negative associations were
found between changes in AUC of hunger (r ˆ À0.87,
P ` 0.05) and PFC (r ˆ À0.84, P ` 0.05) in response Discussion
to the test meal and changes in fasting plasma insulin
for women in response to phase 1 of this program.
Correlation analyses were also performed between The main ®nding of this study is that, in men and
changes in VAS variables in the fasting state and in women having undergone considerable weight loss
response to a meal and changes in body weight and fat through energy restriction followed by a low-fat diet±
mass. These analyses revealed a signi®cant associa- physical exercise intervention, an increase in appetite
tion between changes in fat mass and changes in scores as measured by VAS is observed in the fasting
fasting PFC (r ˆ 0.64, P ` 0.05) in men in response
to phase 1. Table 3 Correlation coef®cients between fasting changes of
leptin, cortisol, insulin and glucose and fasting subjective
feelings of hunger and satiety as measured by VAS during
Correlates of changes of VAS scores in the fasting and phase 2
postprandial states during phase 2 Phase 2
Presented in Table 3 are results from correlation
analyses performed between changes of VAS vari- Insulin Glucose Leptin Cortisol
ables in the fasting state and changes in fasting Men
insulin, glucose, leptin and cortisol which occurred Desire to eat 0.09 0.07 À0.42 0.68*
during phase 2. In men, changes in leptin were Hunger 0.11 0.02 À0.35 0.57
Fullness 0.26 0.02 0.68* À0.78**
positively and signi®cantly associated with changes PFCa À0.07 0.05 À0.33 0.91**
in fullness (r ˆ 0.68, P ` 0.05). In addition, it would Women
seem that in men the change of fasting plasma cortisol Desire to eat 0.41 À0.66 0.59 0.60
Hunger 0.40 À0.63 0.54 0.55
was the most consistent predictor of changes in desire Fullness À0.43 0.71b À0.35 0.13
to eat (r ˆ 0.68, P ` 0.05), fullness (r ˆ À0.78, PFCa 0.21 À0.18 0.37 0.06
P ` 0.01) and PFC (r ˆ 0.91, P ` 0.01). In women, *,**P ` 0.05 and 0.01, respectively.
the best predictor of changes in the variables of VAS a
PFC ˆ prospective food consumption.
would seem to be changes in glycemia, even if the b
P ˆ 0.07.
Table 2 Comparison of VAS measurements calculated as area under the curve in response to a test meal
before and after phase 1 as well as after phase 2
Before phase1 After phase1 After phase 2
Variables Men Women Men Women Men Women
Desire to eat 867 Æ 315 1499 Æ 627 1330 Æ 437 1561 Æ 604 1436 Æ 530 1544 Æ 695
Hunger 956 Æ 334 1546 Æ 621 1379 Æ 469 1494 Æ 615 1506 Æ 535 1532 Æ 701
Fullness 7402 Æ 480 6665 Æ 867 6998 Æ 491 7261 Æ 713 6898 Æ 638 7210 Æ 678
PFCa 1122 Æ 333 2055 Æ 744 1497 Æ 439 1507 Æ 599 1529 Æ 504 1749 Æ 696
Mean ˆ s.e.m. PFC ˆ prospective food consumption; n ˆ 10 men and 7 women, respectively.
International Journal of Obesity

6. Weight loss and appetite
E Doucet et al
911
Figure 2 Correlation analysis between changes in fasting desire to eat and prospective food consumption (PFC) and changes in
fasting cortisol, insulin and leptin in response to an 18-week low-fat diet ± exercise follow-up in obese men (n ˆ 10) and women (n ˆ 7)
having initially undergone 15 weeks of drug therapy.
state. However, a test meal of a given caloric content et al,18 who have found an increase in day long VAS
seems to exert a similar in¯uence over these variables AUC for desire to eat and hunger after 2 and 8 weeks
in the postprandial period once the reduced-obese of energy restriction and physical activity. Further-
state is achieved. Furthermore, the inter-subject varia- more, our results show that these increased feelings of
bility in the the changes of VAS variables in response appetite persist beyond 12 weeks, as demonstrated in
to this program permitted us to investigate the con- the latter study, since our intervention lasted approxi-
tribution of hormonal adaptations to weight loss to mately 33 weeks. These ®ndings are important since
these variations. In this sense, it would seem that appetite feelings might dictate the quantity and the
changes in appetite scores that occur during weight quality of the foods that will be prepared and con-
loss are partly explained by changes in fasting plasma sumed during the day. Thus, it can easily be specu-
cortisol. lated that individuals with increased appetite before
The most consistent ®nding of this study is the the initiation of a meal would rather choose readily
increase of baseline appetite across the intervention in accessible processed foods which generally have a
both men and women. Indeed, as shown in Figure 1, high fat and sucrose content. Moreover, it is also
the desire to eat, hunger and PFC were all increased possible that the amount of food prepared would be
after phase 1 and to a further extent after phase 2. commensurate with the appetite scores. This is in
These results are in accordance with those of Keim agreement with recently reported results which have
International Journal of Obesity

7. Weight loss and appetite
E Doucet et al
912
shown that after an initial weight loss induced over 2 increase to a greater extent in some individuals than in
months by an energy restricted diet, hunger as mea- others during weight loss, it would be expected that
sured by the Dutch Eating Behaviour Questionnaire these particular individuals would lose less fat
scores was a signi®cant predictor of weight regain because of the in¯uence that cortisol seems to exert
over a 14 month period.28 Beyond these observations, over subjective feelings of hunger and satiety and
it would have been expected that, after drug therapy, possibly on food intake.
the low-fat diet±exercise follow-up would have at The recent discovery of the Ob gene product,
least partly corrected for the increase in appetite leptin,31 has brought new insights into the regulation
scores noted after weight loss since the exercise of energy balance. Even if this hormone does not
prescription was of 72% and 52% of VO2 max for seem to be acutely affected by food intake,5,32 it has
men and women, respectively,19 and that high-inten- been shown to ¯uctuate considerably with body
sity exercise seems to reduce ad libitum energy intake weight variations5 and prolonged energy de®cit.33
in the postexercise period in lean healthy men.29 Since leptin has been shown to reduce food intake
However, the fact that an increase in appetite scores in animals,8 it is also probable that it might affect
was still observed during phase 2 is somewhat in energy balance in humans. Even if leptin levels do not
agreement with the observation that the anorectic seem to be associated with hunger and desire to eat
effect of strenuous physical exercise is not necessarily in obese women, be it in the fasting state or after
observed in obese individuals.30 From this, it can be exposure to food,32 other results have demonstrated
argued that the gender difference in exercise intensity that, in response to prolonged energy de®cit induced
which was observed during phase 2 of this weight either by caloric restriction,16 or caloric restriction and
loss±weight maintenance program probably did not aerobic activity,18 changes in hunger±satiety feelings
affect appetite responses differently between men and were related to changes in leptin. Our results do not
women. In this regard, it is also important to consider show as strong a relationship between changes in
that a delay of at least 48 h elapsed between the last leptin and changes in VAS scores. However, our
bout of exercise and VAS measurements, which might results are partly in accordance with those of Heini
be suf®cient to lose a possible acute anorectic effect of et al,16 who reported that leptin seems to be a satiety
exercise. hormone since fullness was signi®cantly associated
Another issue which needs to be discussed is the with changes in leptin in the present study.
fact that, despite a different weight loss, the increases The fact that in women a negative association was
in appetite scores as measured by VAS scale were observed between changes in fasting plasma insulin
quite similar between men and women. It should and changes in variables of VAS measured as AUC is
however be kept in mind that no associations were in accordance with the idea that insulin might affect
observed between changes in body weight and appetite-related variables15 and food intake9 by mod-
changes in appetite-related variables. In this context, ulating the expression of neuropeptide Y.10 Results
it is tempting to speculate that one of the reasons why from a decade ago have also emphasized the impor-
women generally lose less body weight than do men tance of changes in blood glucose in regards to eating
during weight loss programs is because comparable behavior.12,13 More recent results have shed some new
metabolic adaptations to prevent further weight loss light on this issue since Heini et al;15 have demon-
occur in women at a lesser amount of weight lost. This strated that changes in glucose and insulin seem to
®ts well with our results since we document an predict changes in appetite-related variables. Even if
increase in appetite-related variables which is related our data do not provide statistical support to these
to an increase in circulating plasma cortisol more latter results, it nevertheless appears that changes of
particularly during phase 2. Despite a difference in VAS variables also seem to be associated with varia-
changes of body weight and composition, the increase tions in fasting plasma glucose in women.
in the main predictor of changes in appetite-related On the basis of our results, it would seem that the
scores, ie cortisol levels in this case, was remarkably most consistent predictor of changes in appetite-
similar between men and women. Hence, we can related scores, be it at baseline or in response to a
speculate that a lesser weight loss in women than in test meal, is the change in fasting cortisolemia.
men might trigger a similar increase in cortisol levels Indeed, as shown in Figure 2, an increase in fasting
which is in turn associated to similar increases in cortisol seems to better predict an increase of fasting
appetite-related scores more particularly when a resis- desire to eat and PFC when compared to changes in
tance to further loss of fat is achieved. insulinemia or leptinemia. To our knowledge, this is
It is also intriguing to observe a positive relation- the ®rst paper to report such ®ndings in humans
ship between changes in PFC and changes in fat mass having undergone a weight loss intervention program.
in men in response to phase 1. However, this observa- Convincing demonstrations of the effects of glucocor-
tion does not necessarily contradict the main ®nding ticoids on food intake in animals have, however, been
of this study. Indeed, it would seem that the increases reported.34,35 Whether or not these results can be
in appetite scores which occurs during weight loss are transposed to humans remains unclear, but our results
best predicted by increases in cortisol levels and not seem to support observations from the animal litera-
by weight loss per se. In this context, if cortisol levels ture. It is however unclear as to why cortisol comes
International Journal of Obesity

8. Weight loss and appetite
E Doucet et al
913
out as such a strong predictor of changes of VAS 8 Halaas JL, Gajiwala KS, Maffei M, Cohen SL, Chait BT,
variables. Since glucorticoids are involved in glucose Rabinowitz D, Lallone RL, Burley SK, Friedman JM. Weight-
reducing effects of the plasma protein encoded by the obese
homeostasis, and that weight loss is accompanied by gene. Science 1995; 269: 543 ± 546.
important changes in glucose metabolism, changes in 9 Schwartz MS, Figlewicz DP, Baskin DG, Woods SC, Porte D.
cortisol might re¯ect an attempt at re-establishing Insulin in the brain: a hormonal regulator of energy balance.
glucose homeostasis by increasing food intake. Note- Endocrinol Rev 1992; 13: 387 ± 414.
worthy is the fact that an increase in hypothalamic± 10 Schwartz MW, Sipols AJ, Marks JL, Sanacora G, White JD,
Scheurink A, Kahn SE, Baskin DG, Woods SC, Figlewicz DP,
pituitary±adrenal axis activity, as indicated by an Porte D. Inhibition of hypothalamic neuropeptide Y gene
increase in circulating plasma cortisol, might re¯ect expression by insulin. Endocrinology 1992; 130: 3608 ± 3616.
an increase in the level of stress. Even if the role of 11 Wing RR, Klein R, Moss SE. Weight gain associated with
stress in the development of obesity remains to be improved glycemic control in population-based sample of
determined, there is some evidence to the effect that subjects with type I diabetes. Diabetes Care 1990; 13:
1106 ± 1109.
stress can trigger episodes of feeding in animals36 and 12 Camp®eld LA, Smith FJ. Functional coupling between tran-
that some animals display a considerable stress sient declines in blood glucose and feeding behavior: temporal
response when food deprived.37 From these observa- relationships. Brain Res Bull 1986; 17: 427 ± 433.
tions, it might be speculated that energy-restricted 13 Camp®eld LA, Smith FJ. Transient declines in blood glucose
humans might respond similarly to a prolonged signal meal initiation. Int J Obes 1990; 14: 15 ± 31.
14 Weinsier RL, Neslon KM, Hensrud DD, Darnell BE, Hunter
energy de®cit as do animals and, for undetermined GR, Schutz Y. Metabolic predictors of obesity. J Clin Invest
reasons, increase their food intake to alleviate this 1995; 95: 980 ± 985.
stressful status. 15 Heini AF, Kirk KA, Lara-Castro C, Weinsier RL. Relationship
In summary, our results demonstrate that weight between hunger-satiety feelings and various metabolic param-
loss is accompanied by an increase in appetite-related eters in women with obesity during controlled weight loss.
Obes Res 1998; 6: 225 ± 230.
variables in the fasting state. Moreover, based on our 16 Heini AF, Lara-Castro C, Kirk KA, Considine RV, Caro JF,
results the most consistent predictor of these changes Weinsier RL. Association of leptin and hunger-satiety ratings
seems to be changes in fasting plasma cortisol. More in obese women. Int J Obes 1998; 22: 1084 ± 1087.
studies need to be performed to further clarify how 17 Raynaud E, Brun JF, Perez-Martin A, Sagnes C, Boularan
cortisolemia is implicated in the control of energy AM, Fedou C, Mercier J. Serum leptin is associated with the
perception of palatability during a standardized high-carbohy-
balance in the reduced obese state and to what extent drate breakfast test. Clin Sci (Colch) 1999; 96: 343 ± 348.
these changes are acutely affected by food intake. 18 Keim NL, Stern JS, Havel PJ. Relation between circulating
Hence, nutritional strategies must be developed to leptin concentrations and appetite during a prolonged,
prevent the overfeeding which might result from moderate energy de®cit in women. Am J Clin Nutr 1998; 68:
these increased appetite scores, otherwise a very 794 ± 801.
19 Doucet E, Imbeault P, Almeras N, Tremblay A. Physical
plausible scenario is weight regain as is most often activity and low-fat diet: is it enough to maintain weight
seen in reduced obese individuals. stability in the reduced-obese individual following weight
loss by drug therapy and energy restriction? Obes Res 1999;
7: 323 ± 333.
20 Khan MA, Herzog CA, St Peter JV, Hartley GG, Madlon-Kay
Acknowledgements R, Dick CD, Asinger RW, Vessey JT. The prevalence of
This research was supported by grants from Servier cardiac valvular insuf®ciency assessed by transthoracic echo-
Â
Canada and FCAR Quebec. cardiography in obese patients treated with appetite-suppres-
santdrugs. N Engl J Med 1998; 10: 713 ± 718.
21 Weissman NJ, Tighe JFJ, Gottdiener JS, Gwynne JT. An
assessment of heart-valve abnormalities in obese patients
References taking dexfen¯uramine, sustained-release dexfen¯uramine, or
1 Leaf A, Weber PC. A new era for science in nutrition. Am J placebo. Sustained-release dexfen¯uramine study group.
Clin Nutr 1987; 45: 1048 ± 1053. N Engl J Med 1998; 10: 725 ± 732.
2 Blundell J, Burley VJ, Cotton JR, Lawton CL. Dietary fat and 22 Prud'homme D, Langlais M, Samson MP, Gallagher P, Tur-
control of energy intake: evaluating the effects of fat on Â
cotte J, Tremblay A, Despres J-P. Lack of major cardiac
meal size and postmeal satiety. Am J Clin Nutr 1993; 57: valvular abnormalities in asymptomathic obese men and
772S ± 778S. women following a 3-month fen¯uramine or dexfen¯uramine
3 Lawton CL, Burley VJ, Wales JK, Blundell JE. Dietary fat and treatment. Int J Obes 1999; 23: S175.
appetite control in obese subjects: weak effects on satiety. Int J 23 Hill AJ, Blundell JE. The effects of a high-protein or high-
Obes 1993; 17: 409 ± 416. carbohydrate meal on subjective motivation to eat and food
4 Aronne LJ, Mackintosh R, Rosenbaum M, Leibel RL, Hirsch preferences. Nutr Behav 1986; 3: 133 ± 144.
J. Autonomic nervous system activity in weight gain and 24 Arvaniti K, Richard D, Tremblay A. Reproducibility of energy
weight loss. Am J Physiol 1995; 269: R222 ± R225. and macronutrient intake and related substrate oxidation rates
5 Considine RV, Shina MK, Heiman ML, Kriauciunas A, in a buffet-type meal. Br J Nutr (in press).
Stephens TW, Nyce MR, Ohanessian JP, Marco CC, McKee 25 Richterich R, Dauwwalder H. Zur bestimmung der plasma-
LJ, Bauer TL, Caro JF. Serum immunoreactive-leptin concen- glukose-konzentration mit der hexokinase-glucose-6-phos-
trations in normal-weight and obese subjects. N Engl J Med phat-deshydrogenase-methode. Schweiz Med Wochenschr
1996; 334: 292 ± 295. 1971; 101: 615 ± 618.
6 Goldstein DJ. Bene®cial health effects of modest weight loss. 26 Desbuquois B, Aurbach GD. Use of polyethylene glycol
Int J Obes 1992; 16: 397 ± 415. to separate free and antibody-bound peptide hormones in
7 Bray GA. Food intake, sympathetic activity, and adrenal radioimmunoassays. J Clin Endocrinol Metab 1971; 37:
steroids. Brain Res Bull 1993; 32: 537 ± 541. 732 ± 738.
International Journal of Obesity

9. Weight loss and appetite
E Doucet et al
914
27 Couillard C, Mauriege P, Prud'homme D, Nadeau A, Trem- 33 Scholz GH, Englaro P, Thiele I, Scholz M, Klusmann T,
blay A, Bouchard C, Despres JP. Plasma leptin concentrations: Kellner K, Rascher W, Blum WF. Dissociation of serum leptin
gender differences and associations with metabolic risk concentration and body fat content during long term dietary
factors for cardiovascular disease. Diabetologia 1997; 40: intervention in obese individuals. Horm Metab Res 1996; 28:
1178 ± 1184. 718 ± 723.
28 Pasman WJ, Saris WH, Westerterp-Plantenga MS. Predictors 34 Arvaniti K, Deshaies Y, Richard D. Effect of leptin on energy
of weight maintenance. Obes Res 1999; 7: 43 ± 50. balance does not require the presence of intact adrenals. Am J
29 Imbeault P, Saint-Pierre S, Almeras N, Tremblay A. Acute Physiol 1998; 275: R105 ± 111.
effects of exercise on energy intake and feeding behaviour. 35 Dagnault A, Deshais Y, Richard D. Effects of the
Br J Nutr 1997; 77: 511 ± 521. 5-hydroxytryptamine agonist D, L-fen¯uramine on energy
30 Kissileff HR, Pi-Sunyer FX, Segal K, Meltzer S, Foelsch PA. balance in rats: in¯uence of gender. Int J Obes 1993; 17:
Acute effects of exercise on food intake in obese and nonobese 367 ± 373.
women. Am J Clin Nutr 1990; 52: 240 ± 245. 36 Burlet C. Stress and feeding behavior. Ann Endocrinol 1988;
31 Zhang Y, Proenca R, Maffei M, Barone M, Leopold L, 49: 141 ± 145.
Friedman JM. Positional cloning of the mouse obese gene 37 Timofeeva E, Richard D. Functional activation of
and its human homologue. Nature 1994; 372: 425 ± 432. CRH neurons and expression of the genes encoding
32 Karhunen L, Haffner S, Lappalainen R, Turpeinen A, Mietti- CRH and its receptors in food-deprived lean ( Faa?) and
nen H, Uusitupa M. Serum leptin and short-term regulation of obese ( faafa) Zucker rats. Neuroendocrinology 1997; 66:
eating in obese women. Clin Sci 1997; 92: 573 ± 578. 327 ± 340.
Appendix A Composition of the breakfast meal test in men and women
Foods Weight (g) Proteins (g) CHO (g) Fat (g) Energy kJ (kcal)
Wholewheat bread 55 6.9 31.2 1.9 712.(170)
Smooth peanut butter 17 4.3 3.2 8.6 448.(107)
Strawberry jam (Kraft) 20 0.1 14 0 234.(56)
Mozzarella cheese, 17% m.f. 50 13.7 1.6 8.6 582.(139)
Butter 10 0.1 0 8.1 306.(73)
Orange juice 150 1 16.2 0.1 293.(70)
Milk, 2% m.f.a 200 6.7 9.6 3.8 414.(99)
Total women 302 g 26.1 g 66.2 g 27.3 g 2575.(615)
Total men 502 g 32.8 g 75.8 g 31.1 g 2989.(715)
a
Milk was only included in the test meal for men. m.f. ˆ milk fat.
International Journal of Obesity